TRACE ELEMENT ANALYSIS OF MAGNETITE REVEALS LARGE-SCALE OPEN-SYSTEM AND SMALL-SCALE CLOSED-SYSTEM PROCESSES CONTRIBUTE TO MAGNETITE LAYER FORMATION, BUSHVELD COMPLEX, SOUTH AFRICA

The Upper Zone (UZ) of the Bushveld Complex, South Africa, hosts 21 prominent magnetite (mt) ± ilmenite (ilm) and nelsonite (mt + ilm + apatite) layers, and at least five thinner (<10 cm) discrete layers. Layer thickness ranges from 10’s of cm to two meters in the Main Magnetite Layer and nearly ten meters in Layer 21. Several models have been proposed for layer formation, including liquid immiscibility, fractional crystallization, changes in oxygen fugacity and/or pressure, magma recharge and mixing, and double diffusive convection.

Twenty titanomagnetite and adjacent silicate layers from the Eastern Limb and a cm-scale profile through a single ~30 cm thick magnetite layer were studied. Textural observations of significance include mt-exsolution in plagioclase, plagioclase-hosted melt inclusions, and at least two types of symplectite. Mt-exsolution in plagioclase is present throughout most samples, though appears to be less abundant in plagioclase in magnetite layers. Plagioclase-hosted melt inclusions are present in all gabbroic samples in the top third of the UZ, after apatite appears in the mineral assemblage. Symplectites (opx + plag ± cpx, mt primocryst) are present in the lower third of the UZ, disappear, and reappear in samples around the appearance of apatite.

With increasing stratigraphic height, element abundance reversals suggest periodic magma recharge events. Within (thicker) individual layers there is evidence for differentiation, including a general decrease of compatible elements including Cr and V. Plagioclase-hosted melt inclusions and symplectites have been proposed to represent the onset of liquid immiscibility. Melt inclusions and compositional information from proposed conjugate pairs, and the absence of symplectites in samples in the upper portion, suggest that if liquid immiscibility operated, it was only a dominant process in the stratigraphically highest section of the UZ.

Evidence of periodic magma recharge is the most convincing evidence that the UZ, and importantly the magnetite layers, formed in an open system. However, textural and compositional patterns suggest that closed-system processes may have operated at the scale of individual layers and/or packages of silicate layers + magnetite layers.